plotTSNE

Analysis

## Loading required package: ggplot2

## Loading required package: plyr

##   Runing PhenoGraph...

## Run Rphenograph starts:
##   -Input data of 43333 rows and 17 columns
##   -k is set to 30

##   Finding nearest neighbors...DONE ~ 2.316 s
##   Compute jaccard coefficient between nearest-neighbor sets...DONE ~ 16.681 s
##   Build undirected graph from the weighted links...DONE ~ 3.846 s
##   Run louvain clustering on the graph ...DONE ~ 3.885 s

## Run Rphenograph DONE, totally takes 26.728s.

##   Return a community class
##   -Modularity value: 0.863889 
##   -Number of clusters: 23 DONE!
##   Runing t-SNE...with seed 42  DONE

##   Runing PhenoGraph...

## Run Rphenograph starts:
##   -Input data of 138570 rows and 17 columns
##   -k is set to 30

##   Finding nearest neighbors...DONE ~ 9.945 s
##   Compute jaccard coefficient between nearest-neighbor sets...DONE ~ 52.86 s
##   Build undirected graph from the weighted links...DONE ~ 13.499 s
##   Run louvain clustering on the graph ...DONE ~ 15.151 s

## Run Rphenograph DONE, totally takes 91.455s.

##   Return a community class
##   -Modularity value: 0.8810334 
##   -Number of clusters: 33 DONE!
##   Runing t-SNE...with seed 42  DONE

Starting from PBMCs, I think

TSNE results

TSNE results, colored by Phenograph clusters detected

TSNE results, colored by manual gates

TSNE results, Manually gated

Starting from Dendritic, I think

TSNE results

TSNE results, colored by Phenograph clusters detected

TSNE results, colored by manual gates

TSNE results, Manually gated